Here's my take, as a trained computer scientist who has taught college-level students in both computer science and other sciences, and collaborated with scientists at various times: From a practical and pedagogical standpoint, the Excel work you describe sounds like a toy. Training that includes work in a conventional programming language is more likely to prepare these students for several paths:
Being able to work closely with specialists, i.e. computer scientists and/or computational scientists, as collaborators.
Pursuing the option of becoming computational scientists themselves.
Core concepts necessary for scripting at various levels (in Excel, using tools like SciPy, etc.)
Using and truly understanding end-user programming tools like Excel and many other richer environments.
The key reason this type of training is important is twofold: algorithmic thinking and very, very basic software organization. Excel offers a ton of distractions, and masks the semantics and control-flow processes behind its end-user programming model. Adding vbscript on top of the mix just muddies the water.
I would also recommend against starting folks in C++. It has all of the pitfalls of C (which for years made CS programs use Pascal, Modula-2, Lisp, etc. in their first year programs), its own unique pitfalls (see Scott Meyers' and others' series of books on same), with OO stuff poorly stapled on.. and zero useful OO class libraries. (The STL is a generics-based library, not OO!) It is extremely difficult to get students pointed in the right direction early on with C++.
If this does start to appear in the US we should put together a defense pack. And a posse to round up the makes-domesticated-turkeys-look-brilliant patent officials responsible for issuing the damned thing in the first place. What's next, the wheel?
This was also why a lot of folks prefer the competing grsecurity system. First listed among its features (and this has been available in grsec for years):
An intelligent and robust Role-Based Access Control (RBAC) system that can generate least privilege policies for your entire system with no configuration grsec has a lot of other great features; see the link above for details. IMO, it's somewhat unfortunate that grsec has remained a separate patchset for the Linux kernel. Unusable security is useless security; I'm glad to see some catch-up on the SELinux front.
Anyone out there who's used both grsec and SELinux + AppArmour want to favor us with a comparison?
IMO, there's one case where bits are a big win today: I'm sick of real books for technical reference material. It's so much easier to search and link one's way through reference material as a PDF. And when it's out of date (in way less than 20 years) and you need an upgrade, no trees get hurt in the process. If only Adobe would pull its head out of its arse and just copy the search UI of the OS X readers like Preview and (especially) Skim; PDF searching for Windows users seems lacking due to this.
It helps greatly that many publishers have finally figured out how to format PDFs for online readers vs. just chucking camera ready copy (sometimes with registration marks!) at readers.
It doesn't matter to you that medical expenses are the leading cause of bankruptcy in the United States, and for millions of Americans, getting sick or injured at the wrong time can destroy their savings and ruin them for the rest of their lives. Not to mention the litigious bullshit that this induces on the US. Why? Because many folk have NO CHOICE but to attempt to sue to cover their uninsured medical costs, or better still, their insurance company initiates the liability suit on the insured's behalf.. but with no control or say from the insured. Maybe the whole thing is better written off as an accident... instead of suing some elderly person on a fixed income into oblivion.
Travel around parts of Europe for a time, for example. The subtle and not-so subtle attitude changes that come when people aren't deeply afraid of economic debilitation from injury or disease are remarkable. And these changes smack of freedom.
As to the earlier poster's argument about the risk of gov't trying to control your life: a) have you been paying attention to the US political climate? You call this new? and b) that's what the old saw about "eternal vigilance" is for, eh? In this case, it's a matter of the controlling power of corporations (insurance companies) vs. the controlling power of government. At least we have elected voices in one of those groups.
Motion seconded. That was the beauty of the death of the 12" Powerbook line: by having no 12/13" Macbook Pro, it meant that there was no need to artificially restrict the performance or capabilities of the Macbook line. Especially in the first Macbook generation, virtually the only difference aside from exterior parts and display was in integrated vs. discrete graphics.
You're quite right in that ceremony (funeral or otherwise) can be a useful transition tool for the living. But it's not the ceremony per se that many object to, it's how the body of the deceased is dealt with afterwards. Personally, I find using good land to hold onto the preserved corpse of a deceased loved one... wasteful and gruesome. Put another way, casket burial just doesn't scale well.
Yep. What's more, SSD's are just the beginning of a radical shift in the shape of the computer memory hierarchy - the bottom will move "up" considerably. Staggering amounts of key software (operating systems, databases, etc.) are written against the limitations rotating media. SSD's end all of that. No more seek or rotational wait times are just the start. STEC and others will remind us that there's no inherent need to be limited to the read/write speed of the drive head(s) anymore. In simple terms, think of an SSD as "RAID of a mess of flash chips"... but that's a poor analogy, as the hardware connections and software protocols will change and adapt to the newfound performance.
It's that latter implication that's especially disturbing for the Seagates of the world. Almost all hard drive techonology is about to be rendered utterly obsolete, except for a time as protocols for portable "drives". While Seagate litigates, other companies will be working hard on moving past all of the technologies and protocols Seagate is based upon.
It basically says 'GPL your own code as well or STFU'. No, the GPL says 'GPL your own code as well, or write your own damn code.' The very idea that the GPL "removes the freedom of others" is crap. The GPL attempts to stipulate terms which are roughly: share and share alike. Anyone is free to either play by those rules, to take their toys to another sandbox, or to ignore them at the risk of civil action.
is aimed at the "I am incompetent when it comes to technical things and don't understand the concept of URLs"-type people; the like to whom the Internet is the blue IE logo on their desktops. No, it's aimed at people who understand and can leverage search-based interfaces. I freaking love that I can type *just the different/interesting fragment* of a recent/popular URL and typically have FF3 just dredge it up for me. Yes, there's some culture shock when you first use it... but for my purposes its been fast and rockingly useful. As for "awesomebar"... well, we all roll that "1" on the cool-naming die once and awhile.
People are incredibly lazy and only take action when they perceive a threat to their person or property. Liberty? While I understand your frustration, I think it's ultimately misleading to tag people as "lazy" here. Misleading precisely because I tend to agree with you: most folk are "lazy", but that the term is so loaded with negative connotation that it stops further inquiry. Human beings are likey terribly poorly adapted to understanding and reacting to these kinds of threats. Many modern threats are really pretty damn abstract (to an essentially hunter-gatherer mind, anyhow) and require a fair bit of abstract thinking, education, and information to grasp. This presents a huge social risk to be managed: if many modern threats, incidental or deliberate, are difficult for people to properly assess.. our decisions as a society will be distorted dangerously as a result.
I've recently seen reference to work discussing threat perception along these lines, but unfortunately quick searches aren't dredging it up right now. Handy references, anyone?
I don't think it'll happen before we get rid of the limited write cycles Folks, get this through your heads: at the lowest level, NO mass storage media has acceptable reliability anymore. CD's have had error correction built in from the get-go. Magnetic hard drives have been using increasingly sophisticated layers of error correction, bad block marking, etc. for decades now. We will probably never again see a mass storage media that is high-reliability on the basis of a single storage unit (bit or whatever) again; the data densities are likely to just be too low to compete. We'll have a combination of strategies designed to work around the physical limitations of the media (e.g. write-leveling for flash) and layered redundancy/error correction strategies to deal with other failure modes.
What's more: computational processes will probably enter the error-compensation era as well. There have already been articles floating across/. relating to work in this area.
Is it wrong to walk into a gym where you dont have a membership and start exercising just because they dont bother to check ID's at the door? Yes
This is the same thing. It is not wrong to visit a URL. It is wrong to use a pay-service that you are not paying for. No, it isn't the same thing; your analogy is bankrupt. The gym in your example is (presumably) private property, and the unauthorized person who just walks in is a trespasser. The internet is a public space. One cannot trespass in a public space, by definition. Failure to deeply understand the internet as public space leads to all sorts of unsupportable mayhem, such as folks claiming that access to a URL is a "hack", others trying to turn the internet into a "safe, clean space" for their children to wander unattended, etc.
To the original topic: putting a resource on a publicly available URL is an explicit act of publication to the Internet-at-large. I.e. into a vast public forum. The method of discovery of that URL is immaterial -- you could link to it, publish it in the Times, scrawl it on bathroom walls, or let people guess it exists.
If one wants to publish restricted access information via the Internet, then explicit action must be taken to protect it. There are a sea of approaches of varying application and merit, using authentication, encryption, exchange of secret ducks, you name it. But it's the publisher's responsbility to pick one and use it, or else pay for and use a non-public channel.
With Firefox 3.0 beta 3, I upped that to 59/100 by turning off AdBlock Plus on the test page.
Re:People use Photoshop to Dev the Web too Adobe!
on
Adobe To Port AIR To Linux
·
· Score: 2, Insightful
With any such "port my favorite non-OSS app to Linux!" request, two thoughts come to mind:
Is there really a market that would pay for the development and QA effort? In the case of Photoshop, I would suspect that many of those potential users are simply using Mac OS X as their platform of choice these days.
Which release of which distro? You've got to develop and QA against something, and as anyone who has worked with a variety of distros knows, they often just aren't drop-in interchangeable. This question is even more important, as it highlights fragmentation of the Linux desktop userbase. "Linux" doesn't really refer to a single desktop platform target. WINE may help to insulate against some of the lossage here by adapting Linux to a single platform spec (unfortunately, Windows)... but I doubt it'll cover all bases.
Specific to the Adobe Creative Suite apps, what about fully color-managed workflows? Do modern Linux distros have any support for monitor calibration? I'll assume that for soft-proofing and printing, the Adobe apps would handle the print ICC profiles internally as they do on other platforms.
"destroys the market" ?!? Nothing could be further from the truth, IMO.
To date, FOSS is pretty much exactly the software world's implementation of commoditization. In a larger sense, I observe this as the point when that software is no longer interesting solely as a profit center for any one individual (company or geek). It is often, however, rather necessary for some/many folks to get along (e.g. modern computers aren't very useful without an operating system), and thus is a cost center for these users. FOSS helps to distribute these costs, and it has worked brilliantly for many, many people and companies. Yes, some folks no longer get to make money in the that space, but that's what commoditization is all about anyhow! The silver lining comes when others stand upon FOSS' shoulders to build even greater things. Individuals and businesses have been using, creating, and contributing to FOSS projects across the spectrum for some time now to meet their varied needs.
The only unfortunate part in my view, is that no one has yet figured out a hands-down winning structure for open-source software products plus a business model that is a long-term winner over proprietary development for both end-users (RMS' dream) and the developers' financial interests (the pragmatic need). To reiterate: a model where the end-users have the protections afforded by source-code access, and developers (individuals or firms) can make a living and for some to have enough money to afford top-flight R&D and product innovation.
For sony, yes. For end buyers? Nope. To sony this just means their profit margin got bigger. BZZT! Shame on you and the mod that +1 Informative'd you. Does the most blatantly obvious bullet point in the Wii's success story escape you completely? Do try to grasp basic economic reasoning: Sony is out to make more money, but what's really likely given the nature of this product (game console hardware) and how they've been beaten up over their high price point? If possible, they'll implement a price cut to increase their market share. More consoles == more people to sell games to == more profit. Is that really so friggin' hard to grasp?
Heh, I've been making a similar quip/thought-experiment for a long time about having a "de-legislature", the only function of which is to remove law from the books.
But the real problem with both of these ideas is that the existing organizations (legislatures, the USPTO, etc.) really just need to operate for the good of individual citizens, without undue influence by the desires of powerful individuals, organizations, or corporations.
Taking my de-legislature case as an example, it'd be just as bad/good as the original depending on the level/lack of influence by external power influences. A corrupt de-legislature removing laws inconvenient to the powerful would be a pretty awful thing. The same problem applies to a corrupt "office of patent revocation"; it'd just make matters even worse than they already are.
If it is the Cloverfield monster, here's the moral imperative of anyone caught in its landfall zone: SHOOT TO KILL any and all person(s) caught in public with a camcorder but without a steadicam rig...!;-)
Don't sell those users an unlimited data plan then! Here's a clue for you and the mod that marked you "+1 Insightful": Limiting data plans has ZERO effect on a DDOS. None of the individual phones in the scenario described will come anywhere near a bandwidth cap. It's only the conjunction of all phones acting simultaneously that hose the network and/or services on the network.
Ya your right. We should wait to be fucked over, and then react to it. We should wait, you fool. Why? Because there are so many more serious ways we are being fucked over right now that we aren't effectively handling. It's sheer insanity to make up legislation to deal with random useless crap like this. Foresight has its place, but pandering to a manufactured culture of fear is not foresight.
Next on Troll-Attorney-Reality-Show: asshat sues Apple because his iPod is Turing Complete, therefore is capable of doing its owners taxes. "The iPod doesn't do mah taxes, and as we all now know, that makes it crippleware! teh suuxors! teh m0n0p0leeee!"
I hope the judge bats plaintiff and counsel out of the park.
Why in the heck would you want to put fast nonvolatile memory behind a klunky, slow disk interface? Are you nuts? Just map it directly into the system address space and away we go! Use a ramdisk if you really need to work with that storage as a filesystem.
Note that high-speed, high-capacity non-volatile memory completely screws with many built-in assumptions in modern operating systems and the use of their APIs. What happens when a disk orders-of-magnitude slower than RAM no longer slow and isn't even the main system storage? When the implementation of mmap() is radically simpler than read()? The interpretation and implementation of 'persistence' also changes considerably with this flattening of the memory heirarchy, essentially merging the storage capacity and nonvolatility of "mass storage" into the RAM layer.
I was wondering why there can't be a more permanent solution, like creating an image that can be restored using the iTunes Restore function. Most embedded devices with upgradeable firmware have mechanisms to validate new firmware images, e.g. against corruption during the transfer to the device (or downloading before that). Many such devices also decrypt the image and/or verify a cryptographic signature as part of the verification process. Then there's the need to understand the firmware upgrade process sufficiently well so that attempts at a "third-party" upgrade won't just brick the device.
Dealing directly with the firmware upgrade process is generally a bit more involved than the buffer-overflow approach. In some sense, that's the "guarded front gate". Tho I'll note that the buffer-overflow tack might eventually be used to create a back-door that could survive firmware upgrades, even when other installed apps were invalidated by the new firmware.
I call BS. I wrote a rb-appscript tool to sync a master iTunes library (mostly ALAC) to a transcoded (AAC + other) iTunes library, and every last bit of ID3-style metadata was preserved in the files for an entire ~600 album collection. A few dimensions are stored only in the iTunes DB, the user/library specific stuff such as play count, rating, etc.
Note that the aforementioned tool works only by using Applescript to make iTunes transcode files, then transfers those files to the secondary library's directories -- there's no attempt to transfer data directly between the iTunes Library database files. The secondary iTunes picks up everything just fine from only the file-stored metadata.
The possible exception to this may be album art downloaded from iTunes (as opposed to that originally embedded in tracks and/or manually acquired using something like the AmazonArt widget, web search, etc.). Haven't really experimented in this area much yet...
Slashdot Mirror
The key reason this type of training is important is twofold: algorithmic thinking and very, very basic software organization. Excel offers a ton of distractions, and masks the semantics and control-flow processes behind its end-user programming model. Adding vbscript on top of the mix just muddies the water.
I would also recommend against starting folks in C++. It has all of the pitfalls of C (which for years made CS programs use Pascal, Modula-2, Lisp, etc. in their first year programs), its own unique pitfalls (see Scott Meyers' and others' series of books on same), with OO stuff poorly stapled on.. and zero useful OO class libraries. (The STL is a generics-based library, not OO!) It is extremely difficult to get students pointed in the right direction early on with C++.
Anyone out there who's used both grsec and SELinux + AppArmour want to favor us with a comparison?
IMO, there's one case where bits are a big win today: I'm sick of real books for technical reference material. It's so much easier to search and link one's way through reference material as a PDF. And when it's out of date (in way less than 20 years) and you need an upgrade, no trees get hurt in the process. If only Adobe would pull its head out of its arse and just copy the search UI of the OS X readers like Preview and (especially) Skim; PDF searching for Windows users seems lacking due to this.
It helps greatly that many publishers have finally figured out how to format PDFs for online readers vs. just chucking camera ready copy (sometimes with registration marks!) at readers.
Travel around parts of Europe for a time, for example. The subtle and not-so subtle attitude changes that come when people aren't deeply afraid of economic debilitation from injury or disease are remarkable. And these changes smack of freedom.
As to the earlier poster's argument about the risk of gov't trying to control your life: a) have you been paying attention to the US political climate? You call this new? and b) that's what the old saw about "eternal vigilance" is for, eh? In this case, it's a matter of the controlling power of corporations (insurance companies) vs. the controlling power of government. At least we have elected voices in one of those groups.
Motion seconded. That was the beauty of the death of the 12" Powerbook line: by having no 12/13" Macbook Pro, it meant that there was no need to artificially restrict the performance or capabilities of the Macbook line. Especially in the first Macbook generation, virtually the only difference aside from exterior parts and display was in integrated vs. discrete graphics.
You're quite right in that ceremony (funeral or otherwise) can be a useful transition tool for the living. But it's not the ceremony per se that many object to, it's how the body of the deceased is dealt with afterwards. Personally, I find using good land to hold onto the preserved corpse of a deceased loved one... wasteful and gruesome. Put another way, casket burial just doesn't scale well.
Yep. What's more, SSD's are just the beginning of a radical shift in the shape of the computer memory hierarchy - the bottom will move "up" considerably. Staggering amounts of key software (operating systems, databases, etc.) are written against the limitations rotating media. SSD's end all of that. No more seek or rotational wait times are just the start. STEC and others will remind us that there's no inherent need to be limited to the read/write speed of the drive head(s) anymore. In simple terms, think of an SSD as "RAID of a mess of flash chips"... but that's a poor analogy, as the hardware connections and software protocols will change and adapt to the newfound performance.
It's that latter implication that's especially disturbing for the Seagates of the world. Almost all hard drive techonology is about to be rendered utterly obsolete, except for a time as protocols for portable "drives". While Seagate litigates, other companies will be working hard on moving past all of the technologies and protocols Seagate is based upon.
I've recently seen reference to work discussing threat perception along these lines, but unfortunately quick searches aren't dredging it up right now. Handy references, anyone?
What's more: computational processes will probably enter the error-compensation era as well. There have already been articles floating across
This is the same thing. It is not wrong to visit a URL. It is wrong to use a pay-service that you are not paying for. No, it isn't the same thing; your analogy is bankrupt. The gym in your example is (presumably) private property, and the unauthorized person who just walks in is a trespasser. The internet is a public space. One cannot trespass in a public space, by definition. Failure to deeply understand the internet as public space leads to all sorts of unsupportable mayhem, such as folks claiming that access to a URL is a "hack", others trying to turn the internet into a "safe, clean space" for their children to wander unattended, etc.
To the original topic: putting a resource on a publicly available URL is an explicit act of publication to the Internet-at-large. I.e. into a vast public forum. The method of discovery of that URL is immaterial -- you could link to it, publish it in the Times, scrawl it on bathroom walls, or let people guess it exists.
If one wants to publish restricted access information via the Internet, then explicit action must be taken to protect it. There are a sea of approaches of varying application and merit, using authentication, encryption, exchange of secret ducks, you name it. But it's the publisher's responsbility to pick one and use it, or else pay for and use a non-public channel.
With Firefox 3.0 beta 3, I upped that to 59/100 by turning off AdBlock Plus on the test page.
In the case of Photoshop, I would suspect that many of those potential users are simply using Mac OS X as their platform of choice these days.
You've got to develop and QA against something, and as anyone who has worked with a variety of distros knows, they often just aren't drop-in interchangeable. This question is even more important, as it highlights fragmentation of the Linux desktop userbase. "Linux" doesn't really refer to a single desktop platform target. WINE may help to insulate against some of the lossage here by adapting Linux to a single platform spec (unfortunately, Windows)... but I doubt it'll cover all bases.
Specific to the Adobe Creative Suite apps, what about fully color-managed workflows? Do modern Linux distros have any support for monitor calibration? I'll assume that for soft-proofing and printing, the Adobe apps would handle the print ICC profiles internally as they do on other platforms.
"destroys the market" ?!? Nothing could be further from the truth, IMO.
To date, FOSS is pretty much exactly the software world's implementation of commoditization. In a larger sense, I observe this as the point when that software is no longer interesting solely as a profit center for any one individual (company or geek). It is often, however, rather necessary for some/many folks to get along (e.g. modern computers aren't very useful without an operating system), and thus is a cost center for these users. FOSS helps to distribute these costs, and it has worked brilliantly for many, many people and companies. Yes, some folks no longer get to make money in the that space, but that's what commoditization is all about anyhow! The silver lining comes when others stand upon FOSS' shoulders to build even greater things. Individuals and businesses have been using, creating, and contributing to FOSS projects across the spectrum for some time now to meet their varied needs.
The only unfortunate part in my view, is that no one has yet figured out a hands-down winning structure for open-source software products plus a business model that is a long-term winner over proprietary development for both end-users (RMS' dream) and the developers' financial interests (the pragmatic need). To reiterate: a model where the end-users have the protections afforded by source-code access, and developers (individuals or firms) can make a living and for some to have enough money to afford top-flight R&D and product innovation.
Heh, I've been making a similar quip/thought-experiment for a long time about having a "de-legislature", the only function of which is to remove law from the books.
But the real problem with both of these ideas is that the existing organizations (legislatures, the USPTO, etc.) really just need to operate for the good of individual citizens, without undue influence by the desires of powerful individuals, organizations, or corporations.
Taking my de-legislature case as an example, it'd be just as bad/good as the original depending on the level/lack of influence by external power influences. A corrupt de-legislature removing laws inconvenient to the powerful would be a pretty awful thing. The same problem applies to a corrupt "office of patent revocation"; it'd just make matters even worse than they already are.
If it is the Cloverfield monster, here's the moral imperative of anyone caught in its landfall zone: SHOOT TO KILL any and all person(s) caught in public with a camcorder but without a steadicam rig...! ;-)
Next on Troll-Attorney-Reality-Show: asshat sues Apple because his iPod is Turing Complete, therefore is capable of doing its owners taxes. "The iPod doesn't do mah taxes, and as we all now know, that makes it crippleware! teh suuxors! teh m0n0p0leeee!"
I hope the judge bats plaintiff and counsel out of the park.
Why in the heck would you want to put fast nonvolatile memory behind a klunky, slow disk interface? Are you nuts? Just map it directly into the system address space and away we go! Use a ramdisk if you really need to work with that storage as a filesystem.
Note that high-speed, high-capacity non-volatile memory completely screws with many built-in assumptions in modern operating systems and the use of their APIs. What happens when a disk orders-of-magnitude slower than RAM no longer slow and isn't even the main system storage? When the implementation of mmap() is radically simpler than read()? The interpretation and implementation of 'persistence' also changes considerably with this flattening of the memory heirarchy, essentially merging the storage capacity and nonvolatility of "mass storage" into the RAM layer.
Dealing directly with the firmware upgrade process is generally a bit more involved than the buffer-overflow approach. In some sense, that's the "guarded front gate". Tho I'll note that the buffer-overflow tack might eventually be used to create a back-door that could survive firmware upgrades, even when other installed apps were invalidated by the new firmware.
I call BS. I wrote a rb-appscript tool to sync a master iTunes library (mostly ALAC) to a transcoded (AAC + other) iTunes library, and every last bit of ID3-style metadata was preserved in the files for an entire ~600 album collection. A few dimensions are stored only in the iTunes DB, the user/library specific stuff such as play count, rating, etc.
Note that the aforementioned tool works only by using Applescript to make iTunes transcode files, then transfers those files to the secondary library's directories -- there's no attempt to transfer data directly between the iTunes Library database files. The secondary iTunes picks up everything just fine from only the file-stored metadata.
The possible exception to this may be album art downloaded from iTunes (as opposed to that originally embedded in tracks and/or manually acquired using something like the AmazonArt widget, web search, etc.). Haven't really experimented in this area much yet...